Why does it take hours for our users to build up their confidence to approach our products, try to install or use them, consult the manuals in frustration, and finally ask the person down the hall for help? In contrast, why do players of arcade, video, and computer games seem to approach the games without fear, eagerly exploring and learning as they go?

Our research examines the way interface design for applications influences motivation, training, performance, and error correction. In this paper and presentation, we attempt to lay the ground work for additional research by defining the instructional methods used in games and by surveying the literature related to the design of game and business applications.

Introduction: Why study games?
When people sit down to play a game, they are transported to another world where they may get lost or encounter surprises. They usually find this exploration exciting. By contrast, when users get lost in a business application they become frustrated (Carroll, 1982). Games have a way of drawing users to them and keeping them. Perhaps we shouldn’t be surprised that the market for game applications is thriving. Its size and success rivals any other product line in the world. Yet in the business environment, games, and business applications that look like games, often are viewed with disdain. Traditionally, people have viewed leisure activities such as games as a separate activity from work; however, activities we do for fun are not necessarily the opposite of work (Blanchard and Cheska, 1985).

In other words, work involving business applications does not have to be difficult to approach, confusing to learn, or frustrating to use.

In Things That Make Us Smart, Donald Norman identifies seven basic requirements of a learning environment (Norman, 1993):

• Provide a high intensity of interaction and feedback
• Have specific goals and established procedures
• Motivate
• Provide a continual feeling of challenge, one that is neither so difficult as to create a sense of hopelessness and frustration nor so easy as to produce boredom
• Provide a sense of direct engagement, producing the feeling of directly experiencing the environment, directly working on the task
• Provide appropriate tools that fit the user and task so well that they aid and do not distract
• Avoid distractions and disruptions that intervene and destroy the subjective experience

These requirements are satisfied in most games. So why can’t business applications be more like games? They can.

How do people use applications?
As an introduction, consider the following two scenarios of how people use game and business applications:

Scenario 1: Game Application
A person walks through an arcade and her eye is caught by a basketball game called NBA Jam. She’s never seen this game before, but she pauses to watch the game’s attract mode which shows two teams playing hard and fast. They do everything: jump shots, dunks, fancy passes, even half-court three-point shots. As she watches, brief instructions appear on the screen explaining some of the basic strategies and controls of the game. She’s not sure if she’s ready to play, so she waits a minute and two kids put their money in to play. She watches from behind as they play. The kids skip the instructions; instead, they start playing right away. She watches their hands expertly manipulate the controls while their eyes never leave the screen.

After the kids leave, she walks closer to the game. She reads the quick reference instructions next to the controls, then deposits her tokens. More instructions appear on the screen explaining what she has to do to win and providing some basic strategy for playing well. She starts to play. After a minute, the game pauses to tell her she would probably do better if she passed the ball more. It briefly explains how to pass, and within seconds she is playing the game again. At the end of the first period, her coach (the computer) evaluates her performance, providing statistics and hints about how to play more effectively. The computer opponent seems more generous at the beginning, holding back on defense and not trying to block her shots. As she continues playing, the game gets harder, but she barely notices because she’s getting better.

When the game is over, she compares her score to other scores that day. The game keeps track of who she is so it can keep a list of her scores, her favorite team, and the other teams she’s played. The next time she plays, it will know how many games she’s played before -- how many she won, lost, and tied. Depending on the mistakes she makes the next time, she will get different advice. Even though she didn’t win, she enjoyed the few moments of success she had and she believes she’ll do better next time. As she’s standing there watching the attract mode again, learning some new moves, someone else steps up to play. She watches the new player play for a while, then waits eagerly for her turn to play again.

Scenario 2: Business Application
The same person walks down the narrow hallway at her office and sees a friend using a new mail application. It looks interesting, so she asks him where he got it. She goes back to her desk and downloads the mail application from the network. After she installs the

application, she starts it and looks at the interface. She sees some icons at the top and some pulldown menus. The application appears to have a lot of features, but she’s not sure where to start. She tries a few of the icons, but ends up with a screen she doesn’t really want but can’t change.

Next, she tries the online help, which describes the interface components, except she can’t find any information about the icons. She locates the steps for sending a mail message. They’re a little too long for her to remember and she keeps losing the help window behind the mail application, so she prints them. When she gets back from the printer, she starts to follow the steps but realizes that she must already have a mail message to send one, so she goes back to the help. She finds the instructions for how to create a message, but when she tries to create the message, she’s still on the wrong screen and can’t figure out how to get where she’s supposed to be to start following the instructions.

Frustrated, she exits the application and starts it again. Now she’s able to create a message and to send it. After she sends it, she wishes she had kept a copy for herself. She finds out later from a friend that she could have gone into Options and chosen to save a copy of all her sent mail, but it’s too late now. In fact, she finds out later that the application has a lot of useful capabilities she never realized it had.

Her friend stops by and shows her how to turn on the tip of the day feature, which reveals some interesting facts about the application but nothing about confirming sent mail messages. She’s not really sure if her message went through, and she’s not sure if she did it correctly. She exits the application with an uncertain feeling, not sure if she’ll take the time to download any more new applications today and wondering how to get her old mail application back on her PC.

What aspects of games are instructional?
A brief contrast of the two scenarios presented reveals some differences between the instructional methods used in game and business applications as well as perceived influences on the user’s motivation, learning, and performance. The instructional principles reflected in the design of game applications are worth looking at for design ideas, even without knowing quantitatively how much they might improve learning. However, in time, we hope to identify the impact of each method on learning and performance.

Attract Mode
Attract mode is the series of graphics or video that displays on a game when it is not being played. Some modes start on a timed basis or run continuously like a screen saver; others start when a potential player touches a control or moves in front of the game. The purpose of the attract mode is ostensibly to get the attention of a potential user. However, several other invaluable instructional goals are accomplished. The attract mode tells the user about the goal of the game, shows how to use the basic controls, and reveals strategies for winning. This mode also introduces the game’s story, if one exists, which helps immerse the user into the scenario of the game. Many game players watch the attract mode when deciding if they want to play the game, not realizing that while they are deciding they are learning how to play.

In both of the previous scenarios, the new application was perceived as visually appealing, which is an important step toward motivating the user to use the application. However, the game application took the initiative to get the user started. The attract mode demonstrated what could be accomplished with the application and pointed out what the user should do first. The business application tried to reveal all of its features at once without emphasizing any of them, which intimidated the user.

Some business applications are starting to take a more proactive approach to helping the user. For example, Microsoft PowerPoint begins by offering the user a list of options such as open an existing presentation and create a new presentation using wizards. Other Microsoft products provide an optional "Tip of the Day," which provides users with hints about available controls and strategies. However, the majority of applications do not proactively help the user get started.

Business applications very rarely take advantage of the instructional value of the game’s attract mode. Imagine a set of screen savers that appeared after a certain period of inactivity demonstrating model behavior and possible tasks using the tools available on your desktop PC. "Observers learn by watching and imitating others; they tend to behave as they have seen other behave" (Mager, 1984). How often do business applications take the time to reinforce the best way to complete a task? In the current era of increasing features, many designers and developers are spending more time finding multiple ways to accomplish a task rather than taking the time to teach users the best way to accomplish a task.

Clearly Stated Goals
Clearly stated goals are an essential part of effective instruction. "Learners need to know what they are trying to accomplish and how they are doing" (Zemke, 1995).

The game application from the previous scenario had a clearly stated goal that was demonstrated in the attract mode, written on the game casing, and explained in the online instructions shown at the beginning of and often throughout the game. Most users know what they are trying to accomplish before they start the game. Business applications have inherent goals but do not always state them overtly. Moreover, business applications do not attempt to indicate to the user what the goals should be at a particular time. For instance, in the business application scenario, the application could have prompted the user to review and set the mail options immediately after the new installation. Instead, the application left the user at the main screen--full of possibilities but without direction.

Brief Instructions
Although the game application had many features, it didn’t try to tell the user about them all at one time like the business application. Instead, the game provided basic goals and techniques in printed materials near the game controls, and it supplemented those instructions with those that appeared in the attract mode and in the game itself. Also, the game application provided the user with brief instructions at intervals throughout the game, allowing for a gradual and timely assimilation of the information. The business application provided the user with information about each feature, but made the user look for help. It took her out of the application to get the information and interrupted her task.

Transparency
To be successful, learners need to suspend belief and interact with the application on a physical and psychological level without distraction (Laurel, 1991; Low, 1994). Games accomplish this through their use of story and their de-emphasis of the interface controls. A survey of game interfaces shows that games consistently separate the "options" and other "functional" displays from the interactive main display. To view the other display(s), the user simply moves the pointer off the screen or presses a key. By separating the options and other game-level features from the main display, the interface becomes less intimidating and easier to use (Norman, 1993; Bødker, 1990). The user only sees the controls that are currently being used, which keeps them focused on their task. Since there are few controls in the interactive display and they are easy to learn, the user also begins to see a task as a whole, rather than as a series of steps (Betz, 1995; Zemke, 1994).

In the business application scenario, all of the features are accessible from the main window. The user must move the mouse to the specific location for the desired control and click on an icon, button, or menu. As Fitt’s Law suggests, selecting a menu and menu command is much harder to perform than simply moving the mouse off of the screen (Fitt, 1954). In the game, the display matches the current task at hand, so the user is presented only with applicable commands for the current situation. In the business application, an overwhelming number of commands are available, many of which lead to incorrect actions. By attempting to provide complete user freedom, the business application becomes more confusing and discouraging to use.

Performance Coaching
Learning usually requires more than access to information and proper motivation. It requires an element of "the right time and the right place." It requires some kind of guidance that the learners cannot provide themselves. Information or motivation provided too early or too late or even all at once may not result in the most effective learning. In fact, undirected learning could result in bad habits or the creation of improper mental models that must be unlearned before the desired performance can be achieved.

Traditionally, formal guidance for learners comes from teachers. Informally, it can come from asking the person sitting next to you for help or even from observing the behavior of someone next to you. An effective guide helps the learner identify what tasks need to be accomplished; assists the learner in locating the type, level, and amount of information required; encourages the learner, providing hints and suggestions when necessary; and ensures that the learner adopts the optimal performance behavior. Ideally, the guide would be aware of the users’ actions as well as previous user behavior so it could respond in context whenever the user needs help.

In the previous scenarios, only the game application provided the user with coaching tips that told her how to improve her performance. Although rudimentary, the coaching (guidance) provided by the game was aware of the user’s performance. At first, the computer coach provided the user with basic instructions to get started and strategies for winning. During the game, the coach reinforced desired behavior by saying "good shot" and worked to correct undesired behavior by saying "pass" or "run this way." Finally, at the end of the game, the coach gave feedback about the users performance by providing statistics and some parting advice about how to play better the next time.

At best, the business application was able to provide only random tips that might relate to the user’s current situation or something she might want to do eventually. The business application took the approach of "I won’t influence you, the user, because you can do anything you want with me." Unfortunately, the user didn’t know exactly what she could do with the business application- she could not figure out how to get started and she could not tell if she used the tool correctly. Undoubtedly, the user would have benefited from some kind of active guidance that sensed she was in trouble and helped direct her learning.

Instructional designers call this type of guidance "coaching" because the system focuses on aiding optimal performance rather than just providing information (Raybould, 1995). The coach is a vital component of an Electronic Performance Support System (EPSS), which is "an electronic system that provides integrated, on-demand access to information, advice, learning experiences, and tools to enable a high level of job performance with a minimum of support from other people" (Leighton, 1996). EPSS can take many forms, but it includes aids such as wizards and cue cards as well as just-in-time computer-based training. The coach takes a proactive approach to help the user do "whatever is necessary to generate performance and learning at the moment of need" (Gery, 1991).

One study showed that learner control with coaching is more effective than total learner control or browsing (Hannafin, 1992). Another study showed that although learners may complete a task faster with learner control, they do not always have superior retention or recall (Murphy and Davidson, 1991). Other studies have shown that coaching seems to be most effective when the system makes suggestions that the user can ignore (Wynn, 1996; Ross and Morrison, 1988). In other words, a coach should help the user set goals and ensure those goals are achieved effectively; however, the coach should not restrict the users interaction with the application, remove the user from the application, or override the user’s ability to deviate from the norm.

Training Wheels
The game application lets the user be successful from the beginning; the game gets harder as the user gains experience. The business application was difficult from the beginning. As John Carroll’s work with "Training Wheels" interfaces has shown, new users are more successful and are more likely to learn from their errors in a simplified environment (Carroll, 1990). In fact, Carroll’s training wheels research led him to conclude that "incorporating such training wheels into computer systems and applications would produce staged interfaces in which the full complexity of a system could be gradually revealed to users over a course of time and individual experience" (Carroll, 1990).

This description perfectly matches the game environment. The game in the previous scenario reduced the complexity of its features and lowered its difficulty level to allow the user to learn how to perform some basic tasks. By providing a more approachable learning environment, the game increased the user’s confidence and encouraged continued play. Games are always divided into the "staged interfaces" Carroll describes, where each stage becomes a little more difficult and challenging. Many games also provide a training mode that disables more advanced, complex features until the user masters the basic tasks. A business application could provide the same training wheels approach through wizards, cue cards, and modular interactive tutorials.

Consistent Feedback

In the previous scenario, the audio and visual cues in the game application help the user know how well she is progressing towards the goal. Voices and text respond with "Move Faster!" and "Great Move!!!" as she is playing, allowing for quick adjustments and encouraging progression towards the goal state. A continuous score is also displayed for continual assessment. In the game environment, this consistent and continual feedback motivates the user to learn and explore new features. By contrast, the business application seems unaware of the user’s attempt to accomplish a specific goal and provides no encouragement or indication of success or failure. The user is left on her own to determine if her performance was optimal or even correct.

Research has shown that users are most successful with and most receptive to "advisement orientation" feedback (Wynn, 1996; Ross and Morrison, 1988; Murphy and Davidson, 1991; Hannafin, 1992, Carlson, et al., 1992). Advisement orientation combines learner control with advice or coaching from the system. In the game application scenario, the game paused occasionally and offered advice to the player such as "Hold the ‘Run’ button down to run faster." Between game periods, the "coach" would suggest different strategies and moves to explore. In the business scenario, the user was faced with complete "learner control." The system did not help her get started. It did not introduce the application, offer suggestions, or provide any feedback. When the user requested the online help, she was distracted into a separate application that forced her to search for relevant information. When she finally found some of the information she was looking for, it was too overwhelming and took her farther away from her task. To be useful, help information must be integrated into the learning environment. "Tooltips" and the new "What’s This?" help in Windows 95 are a first step in the right direction. The successful learning we see in games suggests that business applications should become much more active and alert to the user’s tasks and goals.

Conclusion
Games have a lot to do with learning. In fact, the activities the user performs playing a game are identical to those required to learn (Norman, 1993). Game and business applications share the same design goals: they should be interactive, challenging, include guided discovery, be relevant and meaningful, and provide appropriate feedback (Hedberg and Harper, 1996; Csikszentmihalyi, 1990; Norman, 1993). By analyzing the instructional methods used in the design of game applications, designers of business applications can gain insight into how to help their users learn while they complete their tasks (Wynn, 1996).

References
Betz, Joseph A. "Computer Games: Increase Learning & Thinking in an Interactive Multidisciplinary Environment." Proceedings of the CIT, 1995.

Betz, Joseph A. "Computer Games: Learning Tools in Engineering & Technology Education." Proceedings of American Society for Engineering Education, Middle Atlantic Conference, 1994.

Blanchard, K. and A. Cheska. The Anthropology of Sport, An Introduction. Bergin & Garvey Publisher, Inc.: Cambridge, 1985.

Bødker, Susan. Through the Interface: A Human Activity Approach to User Interface Design. Lawrence Erlbaum: hillsdale, NJ: 1990.

Carroll, J. M. "The Adventure of Getting to Know a Computer." IEEE Computer, 15(11), 1982: 49-58.

Carroll, J. M. The Nurnberg Funnel: Designing Minimalist Instruction for Practical Computer Skill. MIT Press: Cambridge, 1990.

Csikszentmihalyi, Mihaly. Flow: the Psychology of Optimal Experience. New York: HarperPerennial, 1990.

Fitt, P. M. "The Information Capacity of the Human Motor system in Controlling Amplitude of Movement." Journal of Experimental Psychology, 47, 1954: 381-391.

Gery, G. Electronic Performance Support Systems: How and Why to Remake the Workplace through the Strategic Application of Technology. Boston: Weingarten Publications, 1991.

Hannafin, M. J. "Emerging Technologies, ISD, and Learning Environments: Critical Perspectives." Educational Technology Research and Development, 40(1), 1992: 49-63.

Hedberg, J. and B. Harper. "Interactive Educational Technologies: Effective Design and Application in the Classroom," 3^rd International Interactive Multimedia Symposium, 1996: 160-168.

Laurel, Brenda. Computer as Theatre: A Dramatic Theory of Interactive Experience. Addison-Wesley: Reading, MA: 1991.

Leighton, Chet. "What is an EPSS?" Available from http://itech1.coe.uga.edu/EPSS/Whatis.html.

Low, M. S. Venkataraman, and V. Srivatsan. "Developing an Entrepreneurship Game for Teaching and Research" Simulation & Gaming, 25(3), 1994: 383-401.

Mager, Robert F. Developing Attitude Toward Learning. Belmont, CA: Lake Publishing, 1984.

Murphy, M. A. and G. V. Davidson. "Computer-based Adaptive Instruction: Effects of Learner Control on Concept Learning," Journal of computer-Based Instruction, 18(2), 1991: 51-56.

Norman, Donald A. Things that Make Us Smart: Defending Human Attributes in the Age of the Machine. New York: Addison-Wesley, 1993.

Raybould, B. "Solving Human Performance Problems with Computers," Performance Improvement Quarterly, 8(1), 31-46.

Ross, S. M., and G. R. Morrison. "Adapting Instruction to Learner Performance and Background Variables," In D. H. Jonassen (ed.), Instructional Designs for Microcomputer Courseware, Lawrence Erlbaum Associates: Hillsdale, NJ: 1988, 227-245.

Wynn, Sue. "Interactive Multimedia: Ensuring Motivation of the Learner." Available from http://www.bs.ac.cowan.edu.au/misforum/swynn.html.

Zemke, Ron and Susan Zemke. "Adult Learning: What Do We Know For Sure?" Training, June 1995: 31-40.